<![CDATA[The coastal area of Bedono Village, Demak Regency, Central Java, Indonesia, has experienced substantial land use changes due to the combined impacts of coastal abrasion, tidal flooding, land subsidence, and sea-level rise. Continuous monitoring of these environmental changes is essential to support sustainable coastal management and disaster mitigation. This study aims to analyze multitemporal land use changes between 2015 and 2025 by integrating the Normalized Difference Vegetation Index (NDVI), Normalized Difference Water Index (NDWI), and the Random Forest machine learning algorithm. Landsat 8 and Landsat 9 Level-2 Surface Reflectance imagery were processed using QGIS and the Semi-Automatic Classification Plugin (SCP). NDVI and NDWI were extracted as additional predictor variables to improve spectral separability before image classification. The results indicate a significant reduction in land area accompanied by extensive degradation of mangrove vegetation, particularly during the 2020–2025 period. The Random Forest classification demonstrates that integrating spectral indices substantially improves the identification of complex coastal land cover classes affected by mixed pixels. The resulting multitemporal land use maps provide reliable spatial information for evaluating coastal environmental changes and may support evidence-based decision-making for coastal disaster mitigation, climate change adaptation, and sustainable spatial planning in Demak Regency.]]>
Copyrights © 2026
